- This topic has 12 replies, 5 voices, and was last updated 3 years, 10 months ago by
.
-
Topic
-
I have a 2000 Honda CR-V, automatic, 2WD, American, and I’m pretty sure it’s an LX (I have a hard time keeping it straight, but if it matters, I’ll go check). About a year ago, the timing belt broke (now I know why that’s a maintenance item…) and bent some valves. I took the head off, had it rebuilt, and got everything back on, but I can’t get it to start. I followed Eric’s awesome video “No Spark Diagnoses (Honda)” (https://www.youtube.com/watch?v=9K78XIhhjAQ) which looks like my exact distributor. I show no spark at the plugs, and no spark at the coil, but then when I test the ignitor (ICM) I get an unexpected result. Rather than seeing a ground with the key at ON, that will either flicker at crank (ie, good ignitor, bad coil) or go completely out (bad ignitor), I have no ground whatsoever. I do have a good ground on the distributor body, and I have done the “Ignition Control Module (ICM) Input Test” from page 4-19 of the service manual. This involves unplugging the ICM and making sure all the wires have proper continuity and voltage, even testing the wire from the ICM to the ECM/PCM connector. These point to the ICM being bad, but my impression is that the test assumes a working computer.
So here’s my question. In that video, Eric identifies failure in the ICM by seeing the test light go out completely. My test light doesn’t even come on in the first place. Is that also indicative of a bad ICM, or does that mean I have a problem even earlier than that? If I do have a problem earlier than that, my hunch is that it’s the computer (ECM/PCM or ECU). I pulled that out and opened up the case, and I don’t see any obvious damage or burnt spots. I have not yet done the complete diagnostics for the computer, because I think that my lack of ground at the coil might be definitive.
Any opinions?
-
Replies
-
Oh, yeah, almost forgot, I have no check engine light when I start the car. It’s supposed to go on for two seconds then off, and the ECM/PCM troubleshooting starts with this symptom. So I feel pretty sure that the computer is the problem, but I’d like a little encouragement before I put any more parts in it.
The way the spark is produced is by the igniter (aka ignition module, ICM, etc) switching the ground on and off to the coil primary. The igniter takes the place of the points and condenser on the non-electronic ignition systems. When the ground switches off, the electromagnetic field inside the coil collapses and causes the spark discharge on the coil secondary. This relies on 12v being constantly supplied to the coil when the key is on, and the electronics doing their work on the ground switching.
Do a ‘confidence test’ on the coil. With a voltmeter and the key on, check for the constant 12 volts on the + terminal of the coil’s primary. If you do not have 12v there, you need to stop and find out why before doing anything else. If you have 12v on the coil, now you need to see if the coil has an open or shorted winding. Disconnect all wires to the coil. Using an ohmmeter on its lowest range, check between the + and – terminals on the coil. You should get a very small resistance, usually between 0.4 and 2 ohms. If you have no resistance, it is shorted. If you get infinite resistance, it is opened. If you get the expected resistance, test the secondary. Set the ohmmeter to a higher range and with one of the leads on the primary + terminal, put the other lead where the spark discharges. You should usually read between 6k and 15k ohms. If you get no resistance, infinite resistance, or an unusually low or high resistance, the coil should be replaced. If all is well, then the coil is good.
Ok, so you have done the ‘confidence test’ on the coil, now you need to check if the coil is being switched on and off. A continuity tester or ohmmeter can be used for this. Disconnect the wire at the – side of the coil primary. Connect one side of the tester to a known good ground, and the other side of the tester to the disconnected wire. Now, you or a helper, crank the engine. The tester should show the continuity to ground being switched on and off as the engine turns. If it does, your igniter is good. If the tester stays off or is lit constantly, the igniter is bad.
If the coil and igniter are both good, your problem lies elsewhere.
Is your Honda equipped with an immobilizer feature? If so and this has gone bad, you will not get fire regardless of what you try. If you are not sure, call the dealership with your VIN number…
Thanks for the reply.
Do a ‘confidence test’ on the coil.
I should have mentioned, I did do this, as per page 4-20 in the FSM. The specs in there are 0.63-0.77 ohms for the primary (I got 0.8 ohms, but I’m not willing to deem that a problem until I get the ICM working) and 12.8-19.2 kohms for the secondary. I can’t remember what I had for that, but it was well inside the range.
Ok, so you have done the ‘confidence test’ on the coil, now you need to check if the coil is being switched on and off. Disconnect the wire at the – side of the coil primary. A continuity tester or ohmmeter can be used for this. Connect one side of the tester to a known good ground, and the other side of the tester to the disconnected wire. Now, you or a helper, crank the engine. The tester should show the continuity to ground being switched on and off as the engine turns. If it does, your igniter is good. If the tester stays off or is lit constantly, the igniter is bad.
This is equivalent to the test done in the video, except Eric uses the car battery as the power for his continuity tester. I can do this test, and the tester stays off, but my problem is that the tester was off to begin with. In other words, I think that the test has an unspoken first rule of “the ignitor should provide ground to the coil continually when the key is ON and engine stopped, and switch on and off when the engine cranks.” I get no continuity ever, which leads me to believe this test is not appropriate (yet).
As I’m looking through the manual, I think I can devise a test of the ICM that I’m surprised I haven’t seen online yet. Page 11-72 (terminal 13) shows that the ECM sends 12V to the ICM when the ignition is ON, and pulses when engine is running. So if I manually provide 12V to the ICM, and it does ground the coil, than I have confirmed that the ICM is not getting a signal from the ECM, correct? Granted, this doesn’t completely test the ICM, but at least part of it.
Eric was using a power probe, different than a continuity tester, although the power probe can be used for that purpose. It is a handy tool. I don’t have one, yet…
The coil in nearly all newer cars (I don’t say ALL, because I do not like speaking in absolutes) is ground-side switched. The continuity tester should light (or ohmmeter showing 0.00 or damn close) when there is a path to ground through the igniter. When the igniter switches, the path is broken, so the light goes away (or the ohmmeter shows infinity–or some random high resistance if the switching is faster than the refresh on the meter). An ammeter could also be used for this, showing current to ground when the circuit is closed, and no current when the circuit is switched open, and this would work on both positive and negative sides of the primary. The basic idea is the same; we want to see that some control is happening on the coil (pulsing of the power to the primary side of the coil) to cause secondary discharge.
[quote=”mscherst” post=131020]As I’m looking through the manual, I think I can devise a test of the ICM that I’m surprised I haven’t seen online yet. Page 11-72 (terminal 13) shows that the ECM sends 12V to the ICM when the ignition is ON, and pulses when engine is running. So if I manually provide 12V to the ICM, and it does ground the coil, than I have confirmed that the ICM is not getting a signal from the ECM, correct? Granted, this doesn’t completely test the ICM, but at least part of it.[/quote]
Depending on how the electronics for the igniter were designed, applying 12v directly may not do what you think. In digitally controlled circuits, many are edge based. Meaning, the 12v being applied may not itself be the trigger, but rather 12v being applied and then removed in a prescribed period of time. A pulse of a certain width. Example, 12v being applied for 10ms and then removed. It is the falling edge that is the actual trigger, not just 12v being applied. Additionally, circuits can use the rising edge instead of the falling edge.Also note that many digital circuits use 5v signal/feedback lines, and applying 12v to these parts of the circuit can instantly damage them, frying a formerly good (and potentially expensive) component.
Is your Honda equipped with an immobilizer feature?
Hadn’t thought of that, but I don’t think so. I’ve read a ton of diy guides about this car, and I don’t remember an immobilizer ever being mentioned. A little searching didn’t turn up any confirmed 1st generation CR-Vs with immobilizers, and a couple people saying they were not equipped at all, so I’m going to strongly guess no. Thanks, though.
Depending on how the electronics for the igniter were designed, applying 12v directly may not do what you think. In digitally controlled circuits, many are edge based. Meaning, the 12v being applied may not itself be the trigger, but rather 12v being applied and then removed in a prescribed period of time. A pulse of a certain width. Example, 12v being applied for 10ms and then removed. It is the falling edge that is the actual trigger, not just 12v being applied. Additionally, circuits can use the rising edge instead of the falling edge.
Also note that many digital circuits use 5v signal/feedback lines, and applying 12v to these parts of the circuit can instantly damage them, frying a formerly good (and potentially expensive) component.
Note taken, but does any of this apply in this situation? The direct quote from the manual about the signal from the ECM to the ICM is
“Note: Standard battery voltage is 12V
…
With ignition switch ON (II): battery voltage
With engine running: pulses”Cautions always apply. Plus, other people read these forum posts and apply the advice given to their own situation. No harm in providing as much information as possible.
Does your manual elaborate on the ‘pulses’ as in frequency or modulation of them? I do not know the inner workings of your particular igniter, but some utilize timing circuits and are looking for those pulses to be of a specific duration. Out of range pulses are ignored to protect the igniter. If it is a simple transistor switch, all of the timing of the pulses may be done by the ECM. I do not have that level of technical knowledge about your vehicle. I suppose you can try your improvised testing method, I wouldn’t be able to stop you from doing so anyway. It would seem to me, though, that if what you propose is a viable testing method then it would be outlined in the service manual, especially since it is seemingly simple. There may be a reason why it isn’t there. That unknown reason is why I provided all of the caution.
Cautions always apply. Plus, other people read these forum posts and apply the advice given to their own situation. No harm in providing as much information as possible.
Yes, that’s a good point.
It says nothing about the rate of pulse, but that’s not really what I’m concerned with yet. Even without cranking, I think if the ignition is on, I should get a constant 12V from the ECM.
Thinking about it more, the reason no one has come up with my test yet might be that it doesn’t really tell you anything unless you don’t know what’s going on. My suspicion is that if Eric didn’t have a ground before cranking, he would know that something is wrong coming from the computer. The ignitor failure that he shows starts with a ground, then cuts out completely when cranking. I would guess that when these fail, they can still provide a constant ground but just can’t handle the pulsing. My test would show that the ignitor is still capable of providing at least a constant ground, and that if it’s not doing that with the ignition on, then the ECM isn’t sending a signal. The ignitor might still be bad, but I won’t know until I work out the ECM problem.
I’m really leaning toward finding a used cheap computer. I think the ECM bases the pulse from the TopDeadCenter/CrankshaftPosition/CylinderPosition sensors, and in principle those could go bad, but I don’t know that they would cause a problem with the constant ground before cranking. That and the check engine light issue.
Being old thread so this info won’t help mscherst but may help others. I’m having the same no start problem with a 1998 civic after rebuilding the head and haven’t solved it yet. Anyway, the manual is very poorly written. It is factually correct but severely misleading. The 12 volts on A20 actually comes from the ICM module as I found out afterwards. It is waiting to receive ground to activate its circuit and ground the coil. Way down in the manual (11-34 for me) is a diagram showing A20 (and the injector outputs) connected to NPN transistors such that when the transistor is turned on whatever is on the pin will be grounded. My ECM is not grounding the ICM or the injectors which is (was) probably also the case with mscherst.
I have checked all grounds/voltages/wires several times. I have viewed the CKP/TDC/CYP distributor traces, tried a different ECM and a rebuilt distributor all to no avail. Wonder if mscherst ever solved his problem? These threads seem to die out with no resolution. If I ever get this thing started I will post my solution.
jvf
- You must be logged in to reply to this topic.